DESCRIPTION (Adapted from applicant's description): Roger L. Shapiro, M.D. is an infectious diseases fellow at the Massachusetts General Hospital/Brigham and Women's Hospital program in Boston, and has completed the two year Epidemic Intelligence Service training program at the Centers for Disease Control and Prevention (CDC). His work at CDC was with the Foodborne and Diarrheal Diseases Branch, and his research projects focused on international health issues; he established a diarrheal surveillance system in rural western Kenya, and performed case-control and cohort studies in Kenya and Argentina. His future career plans are to continue working on international epidemiology projects in the field of HIV, through the mentorship of Dr. Max Essex at the Harvard AIDS Institute. HIV-1 prevalence among women of child-bearing age in the southern region of Africa is approximately 25-40%, and more than a third of infants born to these women will become infected. Reliance on breastfeeding throughout Sub-Saharan Africa, and the high prevalence of a potentially more virulent strain of HIV-l (subtype C), make perinatal transmission studies performed in developing countries difficult to generalize to Africa. Unfortunately, areas with the highest HIV-l prevalence are also areas where I not breast-feeding has been associated with higher mortality among infants, largely from diarrheal disease. The protective effect of breast milk among infants born to HIV infected mothers is unknown. An intervention with an extended period of antiretroviral treatment given to infants born to HIV infected women, performed in Botswana where breast-feeding is almost universal and where HIV-1 subtype C predominates, could provide an opportunity to address the unanswered question of a protective role of antiretroviral drugs during breast-feeding in this region of Africa. The. planned intervention will consist of giving: 1) all women ZDV + 3TC from 34 weeks of gestation to one week post-partum, 2) one group of infants ZDV+ 3TC for 1 week followed by formula feeding until 6 months of age, and 3) another group of infants ZDV+3TC for 1 week followed by ZDV and breast-feeding until 6 months of age. The primary objective of the study is to determine the potential efficacy of ZDV for reducing HIV-l subtype C transmission by breast- feeding. In addition to being a co-investigator on this intervention, Dr. Shapiro will be responsible for designing and implementing a diarrheal disease study among this cohort. He will evaluate diarrheal disease among infants through maternal questionnaire data and selected laboratory surveillance; this aspect of the study will be designed to assess the differences between formula and breastfeeding, to study the correlation between maternal viral load and diarrheal disease among the infants, and to determine if maternal HIV infection reduces the protective effect of breastmilk. Ultimately, these findings will further elucidate the risks and benefits of breast feeding among HIV-1 subtype C infected mothers. The study will be carried out in three towns in southeast Botswana as a collaborative effort between the Harvard AIDS Institute and the AIDS/STD Unit of Botswana at Gaborone (AIDSBG).